US10383002B2 - Systems and methods for rapidly estimating available bandwidth in a WiFi link - Google Patents
Systems and methods for rapidly estimating available bandwidth in a WiFi link Download PDFInfo
- Publication number
- US10383002B2 US10383002B2 US15/967,532 US201815967532A US10383002B2 US 10383002 B2 US10383002 B2 US 10383002B2 US 201815967532 A US201815967532 A US 201815967532A US 10383002 B2 US10383002 B2 US 10383002B2
- Authority
- US
- United States
- Prior art keywords
- wifi link
- link
- probe
- wifi
- probe sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
- H04W28/065—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information using assembly or disassembly of packets
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0882—Utilisation of link capacity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0894—Packet rate
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/50—Testing arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/11—Identifying congestion
- H04L47/115—Identifying congestion using a dedicated packet
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/80—Responding to QoS
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/04—Processing captured monitoring data, e.g. for logfile generation
- H04L43/045—Processing captured monitoring data, e.g. for logfile generation for graphical visualisation of monitoring data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
Definitions
- Embodiments described herein relate to estimating WiFi available bandwidth in a WiFi link using frame aggregation.
- WiFi has emerged as a pivotal technology for mobile devices offering the potential for exceptional connectivity speeds.
- Network operators are faced with meeting wireless demands with WiFi ranking as one of the critical technologies to meet those demands.
- WiFi Quality of Experience
- nearly all mobile devices try in some form to push users to WiFi to deliver what is predicted to be a better user Quality of Experience (QoE).
- QoE Quality of Experience
- the general desire is for the network connectivity to simply work.
- Improved cellular connectivity over the past few years has made the choice of cellular versus WiFi more of an economic choice rather than that of a performance choice, particularly with public WiFi. This narrowing of the performance gap has made poor WiFi performance all the more stark in terms of interrupting the seamless connectivity experience for the user.
- WiFi link characterization and more broadly the path characterization
- QoE Quality of Experience
- Test approaches represent Achievable Throughput (AT) tests that test a link by measuring instantaneous Transmission Control Protocol (TCP) throughput.
- TCP Transmission Control Protocol
- this category of tests often take on the order of seconds with significant energy and bandwidth costs, which makes these tests a poor choice for rapid characterization and a sub-optimal choice for longitudinal testing.
- most path characterization techniques tend to be expensive in terms of time, bandwidth, and energy.
- Embodiments described herein address the above shortcomings.
- Systems and methods provided herein use sliced, structured, and reordered packet sequences along with awareness of frame aggregation to rapidly characterize available bandwidth.
- the available bandwidth is characterized within the context of a single web request (for example, a Hypertext transfer protocol (HTTP) GET) method, which consumes limited downlink data with resolution of the path characteristics typically occurring in under a second.
- HTTP Hypertext transfer protocol
- AT Achievable Bandwidth
- Available Bandwidth (AB) testing attempts to discern the residual or remaining capacity on a path or link by analyzing the timing patterns from a sequence of packets.
- adaptive media streaming services such as HTTP live streaming
- MPTCP multipath TCP
- Available Bandwidth tests are often more efficient and arrive more quickly at results as compared to Achievable Throughput tests and with significantly less bandwidth cost.
- the very timing patterns that allow AB tests to operate more efficiently tend to struggle with the increased noise in wireless systems and are not functional under modern WiFi variants (802.11n, 802.11ac) that utilize Frame Aggregation (FA).
- AIWC Aggregation Intensity based WiFi Characterization
- AIWC possesses robustness across varying bandwidth levels, interference, and traffic patterns, and frame aggregation embodies a rich set of WiFi link characteristics with the queuing effects at the specific AP being of particular interest.
- inducing frame aggregation can be induced to measure link congestion and in turn to measure the available bandwidth.
- the concept of Aggregation Intensity (AI) can also be used to capture frame depth, which can be manipulated with targeted packet sequences to capture the likely AB.
- embodiments described here provide a proof of concept AIWC system that realizes the proposed approach through a libpcap-based server with customized TCP flows.
- the methods and systems provided herein operate in-band within TCP to avoid modifications to the end client by leveraging components from TCP (described by S. Savage in 1999 entitled “Sting: A TCP-based Network Measurement Tool,” in USENIX Symposium on Internet Technologies and Systems, vol. 2, incorporated herein by reference; and C. D. Mano, A. Blaich, Q. Liao, Y. Liang, D. A. Cieslak, D. C. Salyers, and A. Striegel, in 2008 entitled “RIPPS” in ACM Transactions on Information and System Security, vol. 11, no. 2, pp. 1-23, incorporated herein by reference).
- the proposed AIWC is also robust across varying bandwidth levels, interference, and traffic patterns, and provides performance improvements versus the existing body of Available Bandwidth literature [(Farshad)].
- Embodiments provided herein explore how the existing frame aggregation mechanisms introduced by 802.11e can be leveraged to achieve such a goal. Examples show not only how frame aggregation breaks existing lightweight mechanisms for link characterization but also how to carefully construct packet sequences that induce frame aggregation to capture the WiFi available bandwidth. Examples show a proof of concept system, AIWC (Aggregation Intensity based WiFi Characterization), to demonstrate the afore-mentioned concepts with significant improvements versus prior work.
- AIWC Aggregation Intensity based WiFi Characterization
- Embodiments described herein include, among other things, a software application which enables users to determine the available bandwidth over a WiFi link using frame aggregation such as, for example, aggregated media access control (MAC) protocol data unit (A-MPDU).
- MAC media access control protocol data unit
- an example embodiment includes an electronic communication device for determining an available bandwidth in an WiFi link using frame aggregation between an access point and the electronic communication device.
- the electronic communication device includes an electronic processor.
- the electronic processor is configured to send a request for a probe sequence.
- the electronic processor is also configured to receive a probe sequence via the WiFi link.
- the electronic processor is also configured to determine an aggregation intensity parameter, the aggregation intensity parameter associated with a number of packets assembled in an aggregated frame for the WiFi link.
- the electronic processor is further configured to determine that an available bandwidth of the WiFi link is less than a probe rate in response to the aggregation intensity parameter being above a threshold.
- Another example embodiment includes a method for determining available bandwidth in a WiFi link using frame aggregation between an access point and an electronic communication device.
- the method includes sending a probe sequence over the WiFi link, wherein the probe sequence includes a plurality of data packets having a fixed packet size and packet gap.
- the method also includes receiving, with the computing device, the probe sequence via the WiFi link.
- the method also includes determining an aggregation intensity parameter, the aggregation intensity parameter associated with a number of packets assembled in an aggregated frame for the WiFi link.
- the method also includes determining that the available bandwidth of the WiFi link is less than a probe rate in response to the aggregation intensity parameter being above a threshold.
- the method also includes displaying, with a display, the available bandwidth of the WiFi link.
- Another example embodiment includes a non-transitory computer-readable medium containing instructions that when executed by one or more electronic processors cause the one or more electronic processors to perform the steps of sending a probe sequence over the WiFi link, wherein the probe sequence includes a plurality of data packets having a fixed packet size and packet gap; receiving, with the computing device, the probe sequence via the WiFi link; determining an aggregation intensity parameter, the aggregation intensity parameter associated with a number of packets assembled in an aggregated frame for the WiFi link; determining that the available bandwidth of the WiFi link is less than a probe rate in response to the aggregation intensity parameter being above a threshold; and displaying, with a display, the available bandwidth of the WiFi link
- one or more devices can be configured to, among other things, conserve resources with respect to power resources, memory resources, communications bandwidth resources, processing resources, and/or other resources while providing mechanisms for controlling and deleting personally identifiable information in content such as documents, audio, and image data.
- conserve resources with respect to power resources, memory resources, communications bandwidth resources, processing resources, and/or other resources while providing mechanisms for controlling and deleting personally identifiable information in content such as documents, audio, and image data.
- the embodiments described herein differ from current technology in the market by providing a quicker technical solution that consumes less bandwidth and energy and is lighter than current technology relating to detecting WiFi link characteristics.
- Technical effects other than those mentioned herein can also be realized from an implementation of the technologies disclosed herein.
- FIG. 1 illustrates an example of a frame format using A-MPDU.
- FIG. 2 is a diagram of a system for estimating WiFi available bandwidth in a WiFi link using frame aggregation in accordance with some embodiments.
- FIG. 3 illustrates the observed receiving packet gap and packet rate in a packet sequence under 802.11g versus. 802.11n.
- FIG. 4 is an example of scheduling under frame aggregation in accordance with some embodiments.
- FIG. 5 illustrates a curve of the probe packet queuing delay versus cross traffic load.
- FIG. 6 illustrates curves of aggregation intensity versus probe packet gap.
- FIG. 7 illustrates curves of aggregation intensity versus cross traffic load with rest to probe packet gap.
- FIG. 8 illustrates AI ⁇ AI base versus cross traffic loaf with respect to the probe rate.
- FIG. 9 shows eCDF of AI ⁇ AI base under uncongested link with respect to the probe rate.
- FIG. 10 illustrates an example probe packet train format with parameters labeled.
- FIG. 11 illustrates curves associated with 802.11g (2.4 GHz) without frame aggregation.
- FIG. 12 illustrates curves associated with 802.11n (2.4 GHz) MIMO with frame aggregation.
- FIG. 13 illustrates curves associated with 802.11ac (5 GHz) MIMO with frame aggregation.
- FIG. 14 illustrates queuing delay observed by PathChirp without (a) and with (b) frame aggregation.
- FIG. 15 shows an available bandwidth result with respect to flow size of TCP cross traffic.
- FIG. 16 illustrates an experimental result from varying cross traffic versus interference traffic.
- FIG. 17 illustrates curves related to AI ⁇ AI base of each sub-train with respect to rate limitation and link utilization.
- FIG. 18 is a flow chart of an example method for estimating available bandwidth in a WiFi link.
- Embodiments described herein disclose how frame aggregation mechanisms can be leveraged to achieve such a goal. Discussed herein is not only how frame aggregation breaks existing lightweight mechanisms for link characterization but also how to carefully construct packet sequences that induce frame aggregation to capture the WiFi available bandwidth. Also discussed herein is a proof of concept system, AIWC (Aggregation Intensity based WiFi Characterization), to demonstrate the afore-mentioned concepts and how the system provides improvements over prior work.
- AIWC Application Intensity based WiFi Characterization
- FIG. 1 illustrates the components of an A-MPDU frame.
- the same principle is applicable to A-MSDU with a different unit (MSDU) as the subframe.
- the two aggregation mechanisms are operated at different levels with the A-MSDU near the top of the MAC layer and the A-MPDU near the bottom of the MAC layer. These two mechanisms may be employed together as specified in 802.11n/ac (described by J. Kolap, S. Krishnan, and N. Shaha, in 2012 entitled “Frame aggregation mechanism for high-throughput 802.11n wlans,” International Journal of Wireless & Mobile Networks ( IJWMN )).
- the two-level aggregation process first needs to wait for the transmission of previous packets in the queue if any; then hold for an additional pre-defined delay to wait for any incoming traffic that is destined for the same address.
- the aggregation is complete if any of the three conditions occur: 1) the size of aggregated frame reaches the maximum ( max ); 2) the estimated transmission time of the aggregated frame reaches the maximum ( max ); or 3) timeout of the pre-defined delay ( ag ).
- Achievable Throughput measures the maximum bandwidth that the TCP flow(s) can achieve by aggressively consuming bandwidth.
- Achievable Throughput estimation exhibits strong intrusiveness by temporarily and partially suppressing existing traffic.
- Achievable Throughput techniques for example, iperf3 or speedtest.net
- iperf3 or speedtest.net tend to be expensive in terms of both time and bandwidth, with costs on the order of tens of seconds and tens of megabytes of data.
- Available Bandwidth is defined as the spare or residual capacity of a link during a time window.
- Available Bandwidth refers to the available bandwidth of the narrow link that has the minimum available bandwidth. The narrow link is different from a tight link where minimal capacity occurs.
- AB techniques are riendlier, preferring to nudge the residual capacity of the bottleneck link rather than consistently competing with existing flows.
- Available Bandwidth techniques accomplish this by creating a series of short traffic bursts that each only last for only a few milliseconds.
- Available Bandwidth methods can be exceptionally lightweight and un-intrusive to the existing traffic.
- the general method to estimate Available Bandwidth is probing from a sender to a receiver. By analyzing the characteristics of received packets with the corresponding sent packets, the available bandwidth on the path from sender to receiver can be approximated.
- PRM Packet Rate Model
- PGM Packet Gap Model
- PRM Packet Gap Model
- the notion of PRM is based on the concept of self-induced congestion. By comparing the probe packet rate with received packet rate, one can detect link congestion if the received packet rate is less than the probe rate, for example, ⁇ ; otherwise, the link is not congested. Thus, the available bandwidth can be discerned by searching the point at which link congestion starts to occur.
- the packet gap and packet rate can be computed with two methods: naive (dividing packet size by packet gap) and jumbo-based idea discussed in Farshad suggested considering aggregated packets as a single jumbo packet for calculating packet rate.
- a jumbo packet can be recognized by finding a series of consecutive packets whose packet gaps under certain threshold (for example, 300 ⁇ s).
- FIG. 2 is a diagram of a system 200 for estimating WiFi available bandwidth in a WiFi link using frame aggregation, in accordance with some embodiments.
- the system 200 includes a server 204 , a network 204 , an access point 206 , and an electronic communications device 210 .
- the WiFi link is established between the access point 206 and the electronic communications device 210 in accordance with protocols specified in the IEEE 802.11 standards.
- the network 204 is a communications network including wireless and wired connections.
- the network 204 may be implemented using a cellular network (for example, a Long Term Evolution (LTE) network), a WiFi network or a combination of both.
- LTE Long Term Evolution
- WiFi Wireless Fidelity
- the concepts and techniques embodied and described herein may be used with networks using other protocols, for example, Global System for Mobile Communications (or Groupe Special Mobile (GSM)) networks, Code Division Multiple Access (CDMA) networks, Evolution-Data Optimized (EV-DO) networks, Enhanced Data Rates for GSM Evolution (EDGE) networks, 3G networks, 4G networks, combinations or derivatives thereof, and other suitable networks, including future-developed networks.
- GSM Global System for Mobile Communications
- CDMA Code Division Multiple Access
- EV-DO Evolution-Data Optimized
- EDGE Enhanced Data Rates for GSM Evolution
- the electronic communications device 210 is a wireless communication device that includes hardware and software that enables the device 210 to communicate via the access point 206 to the network 206 .
- the electronic communications device 210 includes an electronic processor 215 , a memory 220 , an input/output interface 225 , a transceiver 230 , a WiFi processor 235 , a display 240 and an antenna 245 .
- the illustrated components, along with other various modules and components are coupled to each other wirelessly or by or through one or more control or data buses that enable communication there between.
- the electronic processor 215 obtains and provides information (for example, from the memory 220 and/or the input/output interface 215 ) and processes the information by executing one or more software instructions, modules or as part of a program that performs characterization using a modular library or codebase (for example, available bandwidth characterization application 222 ), capable of being stored, for example, in a random access memory (“RAM”) area of the memory 220 or a read only memory (“ROM”) of the memory 220 or another non transitory computer readable medium (not shown).
- the available bandwidth characterization application 222 may include firmware, one or more applications, program data, filters, rules, one or more program modules, and other executable instructions.
- the electronic processor 215 is configured to retrieve from the memory 220 and execute, among other things, software related to the control processes and methods described herein.
- the input/output interface 225 is configured to receive input and to provide output to peripherals.
- the input/output interface 225 obtains information and signals from, and provides information and signals to, (for example, over one or more wires and/or wireless connections) devices both internal and external to the electronic communication device 210 .
- the electronic processor 215 is configured to control the WiFi processor 235 and the transceiver 230 to transmit and receive data to and from the electronic communication device 210 .
- the WiFi processor 235 encodes and the decodes digital data sent and received by the transceiver 230 .
- the transceiver 230 transmits and receives radio signals to and from various wireless communication networks (for example, the network 204 ) using the antenna 245 .
- the electronic processor 215 , the WiFi processor 235 , and the transceiver 230 may include various digital and analog components, which for brevity are not described herein and which may be implemented in hardware, software, or a combination of both. Some embodiments, include separate transmitting and receiving components, for example, a transmitter and receiver, instead of a combined transceiver 230 .
- the display 240 includes a touch screen display, which is a suitable touch-sensitive interface display such as, for example, a liquid crystal display (LCD) touch screens, or an organic light-emitting diode (OLED) touch screen.
- a suitable touch-sensitive interface display such as, for example, a liquid crystal display (LCD) touch screens, or an organic light-emitting diode (OLED) touch screen.
- LCD liquid crystal display
- OLED organic light-emitting diode
- the electronic communications device 210 is a smart telephone. In other embodiments, the electronic communications device 210 may be a table computer, a smart watch, a portable radio, a combination of the foregoing, or another portable or mobile electronic device containing software and hardware enabling it to operate as described herein.
- FIG. 3 shows the observed receiving packet gap and packet rate in a packet sequence under 802.11g vs. 802.11n.
- the dashed line indicates the probe rate.
- FIG. 3 we contrast the impact of FA from 802.11n case ( FIG. 3( b )( d ) ) with the case without FA under 802.11g ( FIG. 3( a )( c ) ).
- the received packet gap under 802.11n FIG. 3( b )
- the reason is that, when frame aggregation is applied, the multiple packets that were assembled in an aggregated frame arrived at the receiver at same time.
- FIG. 3( d ) shows a similar bursty pattern for the naive method.
- PRM when passing through a congested link, the received packet rates should be consistently less than the probe rate as in FIG. 3( c ) .
- this principle breaks with frame aggregation. As shown in FIG. 3( d ) , the received packet rate hovers with over half of the received packets having a packet rate greater than the probe rate. Also, from the perspective of PGM, the majority of compressed packet gaps cannot be used to infer cross traffic anymore.
- Eq (2) can be further extended as:
- T Q is the extra queuing delay resulting from the cross traffic.
- AB Cost of Service
- ⁇ ( ) helps transform the variation of T Q to the impact reflected on the load of cross traffic.
- the relationship between AI and AB is determined by the characteristics of the function ⁇ ( ) between u x and T Q .
- FIG. 4 shows an example of scheduling under frame aggregation.
- the probe traffic competes with the cross traffic
- the cross traffic packets as the A's packets
- the probe packets as the B's packets.
- the function ⁇ ( ) can be approximated as the queuing delay of the probe packets (low-priority) with regarding to the cross traffic (high priority) load.
- Priority queueing systems described by B. D. Choi, D. I. Choi, Y. Lee, and D. K. Sung in 1998 entitled “Priority queueing system with fixed-length packet-train arrivals” in IEEE Proceedings communications; and J.
- Walraevens, S. Wittevrongel, and H. Bruneel in 2007 entitled “A discrete-time priority queue with train arrivals,” in Stochastic Models) have characterized the relationship between packet queuing delay of low-priority packets (i.e., the probe packets) and the traffic load of high-priority packets (i.e., the cross traffic).
- the general pattern of this function can be depicted as the curve in FIG. 5 .
- the curve in FIG. 5 shows the probe packet queueing delay versus cross traffic load.
- the curve in FIG. 5 can divided into two zones: (a) the zone where the probe traffic load plus the cross traffic load is less than the link capacity (the probe rate ⁇ C(1 ⁇ u x )), and (b) the zone where the probe traffic load plus the cross traffic load is greater than link capacity (the probe rate >C(1 ⁇ u x )).
- the increasing slope of the packet delay with the cross traffic load is quite low.
- the queuing delay dramatically rises.
- CBR constant bit rate
- packet gap snd was a large value such that packet gap snd > ag .
- the large gap settings helps gain robustness against wireless dynamics to better utilize the pattern from FIG. 7 .
- snd > ag we intentionally force the aggregation baseline to 1, which means the packets should be separated ideally when no cross traffic is present.
- the aggregation baseline AI base becomes N. Note that this design presumes the cross traffic on wired link(s) would not break the concatenated pattern. As we assume the bottleneck link is the last hop WiFi, the concatenated packets should not be coalesced with the cross traffic on wired link (s).
- FIG. 8 shows the results of the observed AI ⁇ AI base under the different loads of cross traffic where the link capacity is 80 Mb/s. Link congestion occurs when the probe rate plus cross traffic load is greater than the link capacity.
- the observed AI ⁇ AI base matches perfect with the theory of Eq 4. For the high AI base with the high probe rate, the AI can go higher than the lower probe rates.
- FIG. 10 plots the basic format of the probe packet train with the labelled parameters. Table I summarizes the notation and description of the design parameters.
- the first group ( snd , c , tr ) are set in order to satisfy the design principles.
- the second group (M, ) can be tuned by the users for their own goals, e.g., adjusting data cost and tuning estimation resolution.
- snd 1100 ⁇ s to maximize the effectiveness of AI with a minimal time cost.
- c is set to 20 ⁇ s to make sure the intentionally concatenated packets can be aggregated at WiFi link.
- the transient “hiccups” on a sub-train may contaminates the subsequent sub-trains as the temporal queuing effect may not be relieved immediately.
- an inter-sub-train gap tr 2000 ⁇ s to reduce the queuing influence of a sub-train on subsequent sub-trains.
- the setting for the number of packets in a sub-train i is a trade-off between accuracy and cost. A large i can achieve high accuracy with more sample points. It also potentially injects more traffic that may disturb the existing traffic. We recommend setting i proportional to the probe rate which uses more probe packets for the higher probe rate.
- the number of sub-trains M decides the granularity of the test result by determining the searching step
- the algorithm falls back to a conventional approach using the received packet rate to detect link congestion.
- the scoring-based algorithm can be still applied upon the sub-trains.
- AIWC aggregate Intensity based WiFi Characterization
- TCP Sting is used to enable 1:1 ratio of data and ACK packets by leveraging TCP Fast Retransmit.
- the system allows a client to make a web request (HTTP GET) to the server that in turn will serve an AIWC probe packet train as outlined earlier.
- the server takes the responded ACKs to infer the timings of the probes arrived at the client.
- the downstream client itself is unmodified with all intelligence residing at the upstream AIWC server.
- the AIWC client was written using a simple shell script (e.g. curl requests) and was executed on a laptop running Ubuntu 14.04 with multiple WiFi adapter options (Ralink RT3950 for 802.11g/n SISO, EdiMax EW-7822UAC for 802.11n/ac MIMO). The same setting was used for a competing client to generate cross traffic.
- the AIWC client the competing client were connected to a TP-LINK Archer C7 AP (802.11ac capable Access Point) with OpenWrt installed.
- the Access Point was connected through a local Gigabit Ethernet switch to a computer providing NeEM-based (described by S. Hemminger in 2005 entitled “Network Emulation with NetEm,” in Linux Conf Au) emulation for link control.
- the NetEM box was used to emulate many network settings with the tc qdisc, such as delay, packet loss, rate limitation, and the like.
- the cross traffic through the experiment is originated at the server to introduce congestion across the WiFi link.
- Cross traffic was generated by the Distributed Internet Traffic Generator (D-ITG) (described by A. Botta, A. Dainotti, and A. Pescape, in 2012 entitled “A tool for the generation of realistic network workload for emerging networking scenarios,” in Computer Networks) for a fine granularity of control of packets.
- D-ITG Distributed Internet Traffic Generator
- AB iperf3 UDP flow was used to measure the WiFi capacity. Then AB can be calculated by subtracting the throughput of cross traffic from the link capacity.
- PathChirp is a typical PRM (Packet Rate Model)
- Spruce is a typical PGM (Packet Gap Model)
- WBest+ is aware of frame aggregation and is particularly designed for WiFi. All three methods were configured as specified in their respective papers (for example, Spruce requires being informed of the bottleneck capacity).
- FIGS. 11, 12 and 13 we plot the results relating to the ground truth for each of the methods. Note that the diagonal dashed line in the figures indicates the optimal case where the estimation result is equal to the ground truth.
- AIWC and WBest+ show better performance. Since Spruce and PathChirp are designed specifically for wired links with targets of large bandwidths on the order of hundreds of megabits, the approaches struggle to sense the subtle AB variations on the WiFi link. For AIWC, when consistently low AI was detected, AIWC adopts the backward compatible algorithm mentioned in Section IV-C.
- AIWC achieves a fine granularity with a resolution of 2 Mb/s that helps outperform WBest+.
- the normal fluctuations of WiFi link capacity makes the AB in this low region difficult to be consistent.
- the flow size will decide the time fraction of the link being saturation.
- the WiFi link remains largely unconsumed which leads to 80% of the AB test returning >100 Mb/s for a 1 KB flow size.
- the flow size increases, the AB result moves towards the lower regions.
- 60% of tests returns AB ⁇ 10 Mb/s, which implies that the link was largely saturated.
- the result shows AIWC can effectively reflect the impact of TCP traffic upon AB of the link.
- AB on WiFi is decided not only by the competing traffic on the same AP but also by the interference on the same or overlapped channel.
- We set up another AP (interference AP) to run on the same channel as the major AP in order to generate interference.
- By varying the traffic on the interference AP we are able to vary the AB on the link between the major AP and the AIWC client. Similar to the case of cross traffic, we take the AB ground truth as the link capacity minus the throughput of interference traffic.
- rate limiting in public WiFi (for example, public guest WiFi), it was important to understand how AIWC would react to rate limiting.
- token bucket we employed the tc qdisc htb tool to cap the maximum rate of outbound traffic of the AP.
- rate limits we adopted the experiment setting of the cross traffic to vary the link utilization. The experiment was conducted under an 802.11ac link.
- FIG. 17 plots the AI ⁇ AI base for each sub-train in a (10, 100) Mb/s train. Without rate limiting, we can see the AI ⁇ AI base is consistently above zero, and the cross traffic can push this value higher by forcing more packets to be aggregated. With rate limiting, once a sub-train i with probe rate R; is greater than the throttled rate, the AI ⁇ AI base drops negative, which implies the aggregated packets (according to the concatenating approach) were torn apart. For example, under a rate limit of 20 Mb/s, the AI ⁇ AI base of the sub-trains whose probe rate is larger than 20 Mb/s start to go negative because the rate limitation prevents the high rate sub-trains from concatenating more packets to generate high probe rates.
- the observed AI stays low while the AI base increases with the probe rate.
- the cross traffic and the probes share the tokens of rate limiting, the cross traffic can make the negative pattern occur on even lower rate sub-trains. This observation implies that AIWC can recognize the rate limiting by detecting if the AI ⁇ AI base goes consistently negative.
- FIG. 18 is a flow chart of an example method for estimating available bandwidth in a WiFi link.
- the server 202 is configured to send a probe sequence over the WiFi link connecting the access point 206 and the electronic communication device 210 .
- the probe sequence (as shown in FIG. 10 ) is generated at the server 202 .
- the probe sequence includes a plurality of data packets. In one example, the data packets have a fixed packet size and packet gap between consecutive data packets in the probe sequence.
- the electronic communication device 210 sends a HTTP GET request to the server 202 .
- the web request from the electronic communication device 210 further includes a request for a target bit rate or other parameters to select appropriate modes of operation.
- the electronic communication device 210 is capable of choosing the highest speed to search for (for example, requests the server 202 to test for 10 Mb/s).
- the electronic communication device 210 is configured to receive the probe sequence from the access point 206 , via the WiFi link.
- the electronic communication device 210 determines an aggregation intensity parameter associated with a number of packets assembled in an aggregated frame for the WiFi link.
- the aggregated frames used for the WiFi link is in accordance with A-MPDU as specified in the IEEE 802.11 standards.
- the electronic communication device 210 in response to the aggregation intensity parameter determined in block 330 being above a threshold (AI con , as discussed earlier) from the baseline value AI base as shown in the Eq (4) above, the electronic communication device 210 determines an available bandwidth of the WiFi link is less than a probe rate
- the electronic communication device 210 is configured to display the available bandwidth of the WiFi link.
- the embodiments described here may be used to rank available WiFi connections, which may be used to automatically select and connect to a WiFi connection likely providing the best QoE (for example, the optimal WiFi connection with the least current user traffic).
- the ranked WiFi connections may be displayed to a user and the user may be allowed to select a WiFi connection to connect to.
- the displayed rankings As compared to other rankings that may display WiFi connection rankings based on potential peak performance, the displayed rankings, determined using the above systems and methods, display actual usable capacity that depends on current user traffic.
- embodiments described herein deliver a QoE rating in terms of quantified A-MPDU levels as compared to existing methods of displaying signal strength in terms of bars.
- some embodiments described herein use passive detection so users do not have to manually initialize a test each time they want to reassess available WiFi connections.
- the embodiments described herein allows a client to access information the loading on an access point to assist in the selection of an optimal WiFi link. Some embodiments also allow this process to be carried out with almost no additional overhead. With access to such detailed information about an access point, the client is able to rank the best networks available and choose (automatically or based on manual selection) which access point to connect to accordingly.
- the embodiments described herein may provide users with information the A-MPDU and rate levels, so that the best network can be chosen.
- the embodiments described herein may present the load of each network to the user through access point side reporting of the A-MPDU levels of outbound packets.
- the client is able to do this by way of a promiscuous scan during a normal Beacon scanning process that may take place during the course of a standard WiFi connection.
- these embodiments allow a client to simultaneously and continuously listen for responses from available access points as well as viewing the resulting A-MPDU and rate levels of the detected access points.
- the embodiments described here measure lading on 802.11 WiFi network.
- the embodiments may use passive technology and can be built into functional app versions (for example, Android or iOS) that periodically scan (for example, continuously scan 24 hours a day, 7 days a week). Accordingly, some embodiments require zero client modification. In other words, some embodiments work within existing protocols and protocol stacks, such as TCP/IP and HTTP GET.
- Embodiments described herein also detect loading fast with some embodiments having a resolution time of under 250 milliseconds. Embodiments also provide accurate loading measurements or rankings and may designate between red, yellow, and green rankings within 0-11 Mb/s. Embodiments are also lightweight, as some embodiments carry less than 100 KB of data.
- Embodiments described herein may also be used in secondary applications. For example, embodiments may be used for cellular adaptation to handle larger aggregation windows. Embodiments may also be used for cloud adaptation, adaptive probing, actionable quality of experience, including both historical and geographic actions, or a combination thereof.
- Articles “a” and “an” are used herein to refer to one or more than one (i.e. at least one) of the grammatical object of the article.
- an element means at least one element and can include more than one element.
- embodiments described herein may include hardware, software, and electronic components or modules that, for purposes of discussion, may be illustrated and described as if the majority of the components were implemented solely in hardware.
- electronic based aspects of the invention may be implemented in software (for example, stored on non-transitory computer-readable medium) executable by one or more processors.
- mobile device and “computing device” as used in the specification may include one or more electronic processors, one or more memory modules including non-transitory computer-readable medium, one or more input/output interfaces, and various connections (e.g., a system bus) connecting the components.
Abstract
Description
In summary, it can be seen that both of the methods heavily rely on the fact that, the received packet rate or packet gap should be an effective indicator to reveal the temporal cross traffic. The observations rely on packets being scheduled in a first-in-first-out (FIFO) manner and on a per-packet basis.
C. Available Bandwidth Estimation Under the Impact of Frame Aggregation
subject to
When the probe traffic exceeds the current AB and triggers link congestion, the AI would enter into the second zone and increase dramatically. Therefore, by simply checking if the observed AI increases by a certain threshold (AIcon) from the baseline value AIbase, we are able to tell if the AB of a WiFi link is less than the probe rate
The expression can be summarized as follows:
In this case, the aggregation baseline AIbase becomes N. Note that this design presumes the cross traffic on wired link(s) would not break the concatenated pattern. As we assume the bottleneck link is the last hop WiFi, the concatenated packets should not be coalesced with the cross traffic on wired link (s).
TABLE I |
PROBE PACKET TRAIN DESIGN PARAMETERS DESCRIPTION |
Gsnd | The packet gap between non-concatenated packets | ||
Gc | The packet gap between concatenated packets | ||
Gtr | The packet gap between neighbour sub-trains | ||
M | The total number of sub-trains | ||
Li | The number of packets in the i-th sub-train | ||
Pi | The packet size for the i-th sub-train | ||
Ri | The probe rate of the i-th sub-train | ||
Ni | The number concatenated packets to achieve Ri | ||
B. Parameter Setting
Given a certain i a large M offers good estimation resolution but at expensive data cost.
C. Estimation Algorithm
to sub-train i; otherwise, a same value of negative score is added. By assigning the weight as reciprocal of sub-train index difference, we value the consensus of the closely transmitted sub-trains more than of sub-trains far apart. Finally, AB can be approximated with the probe rate of the sub-train with the highest score. If AB is not in the range of [Rmin, Rmax], the result will be given as a classification: 1) if all sub-trains have AB<Ri, then it returns AB<Rmin; 2) if AB>Ri observed for all sub-trains, it returns AB>Rmax.
TABLE II |
COST COMPARISON ACROSS DIFFERENT METHODS |
Method | Time (s) | Traffic (KB) | Pkts # |
AIWC | (1, 20) | 0.18 | 167.83 | 210 |
(Rmin, Rmax) | (10, 80) | 0.10 | 314.10 | 260 |
(10, 100) | 0.10 | 373.14 | 300 |
Wbest+ | 0.95 | 405.14 | 309 |
Spruce | 19.16 | 766.24 | 734 |
PathChrip | 10.36 | 305.21 | 222 |
iperf3 (throughput test) | 10 | 11,851.90 | 8,940 |
At
Claims (17)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/967,532 US10383002B2 (en) | 2017-05-01 | 2018-04-30 | Systems and methods for rapidly estimating available bandwidth in a WiFi link |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201762492397P | 2017-05-01 | 2017-05-01 | |
US201762492578P | 2017-05-01 | 2017-05-01 | |
US15/967,532 US10383002B2 (en) | 2017-05-01 | 2018-04-30 | Systems and methods for rapidly estimating available bandwidth in a WiFi link |
Publications (2)
Publication Number | Publication Date |
---|---|
US20180317131A1 US20180317131A1 (en) | 2018-11-01 |
US10383002B2 true US10383002B2 (en) | 2019-08-13 |
Family
ID=63917004
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/967,532 Active US10383002B2 (en) | 2017-05-01 | 2018-04-30 | Systems and methods for rapidly estimating available bandwidth in a WiFi link |
Country Status (1)
Country | Link |
---|---|
US (1) | US10383002B2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11025524B2 (en) * | 2018-08-20 | 2021-06-01 | Exfo Inc. | Telecommunications network and services QOE assessment |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114390570B (en) * | 2020-10-20 | 2023-11-21 | 中国移动通信集团北京有限公司 | Network rate detection method, device, equipment and computer storage medium |
Citations (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020181394A1 (en) * | 2000-08-31 | 2002-12-05 | David Partain | Bandwidth broker for cellular radio access networks |
US20040039817A1 (en) | 2002-08-26 | 2004-02-26 | Lee Mai Tranh | Enhanced algorithm for initial AP selection and roaming |
US20060252443A1 (en) | 2005-04-25 | 2006-11-09 | Interdigital Technology Corporation | Method and system for efficient addressing and power savings in wireless systems |
US20080102852A1 (en) | 2006-10-25 | 2008-05-01 | Lei Du | Method and Apparatus for Access Point Selection in Wireless LAN |
US20090013210A1 (en) * | 2007-06-19 | 2009-01-08 | Mcintosh P Stuckey | Systems, devices, agents and methods for monitoring and automatic reboot and restoration of computers, local area networks, wireless access points, modems and other hardware |
US7573857B1 (en) | 2004-01-16 | 2009-08-11 | Qualcomm Incorporated | Capacity management for wireless local area networks |
US20090303871A1 (en) | 2008-06-10 | 2009-12-10 | Electronics Telecommunications Research Institute | Method and apparatus for packet aggregation according to traffic characteristics |
US7864736B2 (en) | 2007-09-24 | 2011-01-04 | Tcm Mobile Llc | Packet communication roaming method and system |
US20120294181A1 (en) * | 2006-06-09 | 2012-11-22 | Svante Ekelin | Data transfer path evaluation using filtering and change detection |
US20130090517A1 (en) * | 2011-10-07 | 2013-04-11 | Cochlear Limited | Flexible Protocol for an Implanted Prosthesis |
US20130142095A1 (en) | 2011-12-02 | 2013-06-06 | Futurewei Technologies, Inc. | System and Method for Traffic Signaling and Control in a Wireless Network |
US20130329620A1 (en) | 2011-01-16 | 2013-12-12 | Lg Electronics Inc. | Method for communication based on identifying information assignment and apparatus for the same |
WO2014023337A1 (en) | 2012-08-07 | 2014-02-13 | Nokia Siemens Networks Oy | Mechanism for controlling discovery and selection function of access points |
US20150036673A1 (en) | 2013-07-30 | 2015-02-05 | Qualcomm Incorporated | Systems and methods for communicating multi-destination traffic in a wireless network |
US20150063111A1 (en) | 2013-08-28 | 2015-03-05 | Qualcomm Incorporated | Methods and apparatus for multiple user uplink |
US8988999B2 (en) | 2012-05-30 | 2015-03-24 | Intel Corporation | Method, system and apparatus of wireless local area network (WLAN) communication in conjunction with cellular communication |
US20150146648A1 (en) | 2013-11-25 | 2015-05-28 | Canon Kabushiki Kaisha | Method and device for data communication in a communication network |
US20150289201A1 (en) | 2014-04-07 | 2015-10-08 | Qualcomm Incorporated | Method and apparatus for an improved mechanism for selecting an access point |
US20150351026A1 (en) | 2014-05-27 | 2015-12-03 | Samsung Electronics Co., Ltd | Method and apparatus for scanning access point in wireless communication system |
US20150350974A1 (en) | 2014-05-30 | 2015-12-03 | Qualcomm Incorporated | Enhancement for bss transition, load balancing and ap selection |
US20160066227A1 (en) | 2013-03-28 | 2016-03-03 | British Telecommunications Public Limited Company | Access point selection in a wireless network |
CN105491642A (en) | 2015-11-25 | 2016-04-13 | 小米科技有限责任公司 | Network connection method and device |
US20160112984A1 (en) | 2014-10-16 | 2016-04-21 | Qualcomm Incorporated | Communication between devices of a neighbor aware network |
US20160142163A1 (en) | 2012-12-03 | 2016-05-19 | Intel Corporation | Apparatus, system and method of user-equipment (ue) centric access network selection |
CN105657760A (en) | 2016-03-01 | 2016-06-08 | 华中科技大学 | WiFi seamless switching method based on neighbor graph algorithm |
CN106487476A (en) | 2015-09-01 | 2017-03-08 | 华为技术有限公司 | The reception state indicating means of A-MPDU and receiving device |
US20170111854A1 (en) | 2015-10-15 | 2017-04-20 | Qualcomm Incorporated | Quality of service aware access point and device steering |
US20170111821A1 (en) | 2015-10-19 | 2017-04-20 | Relay2, Inc. | Distributed load balancing for access points |
US20170111951A1 (en) | 2015-10-20 | 2017-04-20 | Marvell World Trade Ltd. | Acknowledgment Data Unit for Multiple Uplink Data Units |
US20170181045A1 (en) | 2014-08-18 | 2017-06-22 | Hewlett Packard Enterprise Development Lp | Selecting a subset of access points |
US20170288996A1 (en) * | 2014-08-28 | 2017-10-05 | British Telecommunications Public Limited Company | Network connectivity identification |
US9788269B2 (en) | 2015-03-20 | 2017-10-10 | Qualcomm Incorporated | Selection of an access point in a wireless communications network |
US20180034740A1 (en) * | 2015-02-11 | 2018-02-01 | Telefonaktiebolaget Lm Ericsson (Publ) | Ethernet congestion control and prevention |
US20180063748A1 (en) * | 2016-09-01 | 2018-03-01 | Alcatel-Lucent Usa Inc. | Estimating bandwidth in a heterogeneous wireless communication system |
US20180092111A1 (en) * | 2016-09-28 | 2018-03-29 | Qualcomm Incorporated | Maximizing a frame's arrangement thereby increasing processing time available to processors |
US10164867B2 (en) * | 2017-02-03 | 2018-12-25 | Cisco Technology, Inc. | Generating non-congruent paths having minimal latency difference in a loop-free routing topology having routing arcs |
-
2018
- 2018-04-30 US US15/967,532 patent/US10383002B2/en active Active
Patent Citations (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020181394A1 (en) * | 2000-08-31 | 2002-12-05 | David Partain | Bandwidth broker for cellular radio access networks |
US20040039817A1 (en) | 2002-08-26 | 2004-02-26 | Lee Mai Tranh | Enhanced algorithm for initial AP selection and roaming |
US7573857B1 (en) | 2004-01-16 | 2009-08-11 | Qualcomm Incorporated | Capacity management for wireless local area networks |
US20060252443A1 (en) | 2005-04-25 | 2006-11-09 | Interdigital Technology Corporation | Method and system for efficient addressing and power savings in wireless systems |
US20120294181A1 (en) * | 2006-06-09 | 2012-11-22 | Svante Ekelin | Data transfer path evaluation using filtering and change detection |
US20080102852A1 (en) | 2006-10-25 | 2008-05-01 | Lei Du | Method and Apparatus for Access Point Selection in Wireless LAN |
US20090013210A1 (en) * | 2007-06-19 | 2009-01-08 | Mcintosh P Stuckey | Systems, devices, agents and methods for monitoring and automatic reboot and restoration of computers, local area networks, wireless access points, modems and other hardware |
US7864736B2 (en) | 2007-09-24 | 2011-01-04 | Tcm Mobile Llc | Packet communication roaming method and system |
US20090303871A1 (en) | 2008-06-10 | 2009-12-10 | Electronics Telecommunications Research Institute | Method and apparatus for packet aggregation according to traffic characteristics |
US20130329620A1 (en) | 2011-01-16 | 2013-12-12 | Lg Electronics Inc. | Method for communication based on identifying information assignment and apparatus for the same |
US20130090517A1 (en) * | 2011-10-07 | 2013-04-11 | Cochlear Limited | Flexible Protocol for an Implanted Prosthesis |
US20130142095A1 (en) | 2011-12-02 | 2013-06-06 | Futurewei Technologies, Inc. | System and Method for Traffic Signaling and Control in a Wireless Network |
US8988999B2 (en) | 2012-05-30 | 2015-03-24 | Intel Corporation | Method, system and apparatus of wireless local area network (WLAN) communication in conjunction with cellular communication |
WO2014023337A1 (en) | 2012-08-07 | 2014-02-13 | Nokia Siemens Networks Oy | Mechanism for controlling discovery and selection function of access points |
US20160142163A1 (en) | 2012-12-03 | 2016-05-19 | Intel Corporation | Apparatus, system and method of user-equipment (ue) centric access network selection |
US20160066227A1 (en) | 2013-03-28 | 2016-03-03 | British Telecommunications Public Limited Company | Access point selection in a wireless network |
US20150036673A1 (en) | 2013-07-30 | 2015-02-05 | Qualcomm Incorporated | Systems and methods for communicating multi-destination traffic in a wireless network |
US20150063111A1 (en) | 2013-08-28 | 2015-03-05 | Qualcomm Incorporated | Methods and apparatus for multiple user uplink |
US20150146648A1 (en) | 2013-11-25 | 2015-05-28 | Canon Kabushiki Kaisha | Method and device for data communication in a communication network |
US20150289201A1 (en) | 2014-04-07 | 2015-10-08 | Qualcomm Incorporated | Method and apparatus for an improved mechanism for selecting an access point |
US20150351026A1 (en) | 2014-05-27 | 2015-12-03 | Samsung Electronics Co., Ltd | Method and apparatus for scanning access point in wireless communication system |
US20150350974A1 (en) | 2014-05-30 | 2015-12-03 | Qualcomm Incorporated | Enhancement for bss transition, load balancing and ap selection |
US20170181045A1 (en) | 2014-08-18 | 2017-06-22 | Hewlett Packard Enterprise Development Lp | Selecting a subset of access points |
US20170288996A1 (en) * | 2014-08-28 | 2017-10-05 | British Telecommunications Public Limited Company | Network connectivity identification |
US20160112984A1 (en) | 2014-10-16 | 2016-04-21 | Qualcomm Incorporated | Communication between devices of a neighbor aware network |
US20180034740A1 (en) * | 2015-02-11 | 2018-02-01 | Telefonaktiebolaget Lm Ericsson (Publ) | Ethernet congestion control and prevention |
US9788269B2 (en) | 2015-03-20 | 2017-10-10 | Qualcomm Incorporated | Selection of an access point in a wireless communications network |
CN106487476A (en) | 2015-09-01 | 2017-03-08 | 华为技术有限公司 | The reception state indicating means of A-MPDU and receiving device |
KR20180026518A (en) | 2015-09-01 | 2018-03-12 | 후아웨이 테크놀러지 컴퍼니 리미티드 | A-MPDU < / RTI > |
US20170111854A1 (en) | 2015-10-15 | 2017-04-20 | Qualcomm Incorporated | Quality of service aware access point and device steering |
US20170111821A1 (en) | 2015-10-19 | 2017-04-20 | Relay2, Inc. | Distributed load balancing for access points |
US20170111951A1 (en) | 2015-10-20 | 2017-04-20 | Marvell World Trade Ltd. | Acknowledgment Data Unit for Multiple Uplink Data Units |
CN105491642A (en) | 2015-11-25 | 2016-04-13 | 小米科技有限责任公司 | Network connection method and device |
CN105657760A (en) | 2016-03-01 | 2016-06-08 | 华中科技大学 | WiFi seamless switching method based on neighbor graph algorithm |
US20180063748A1 (en) * | 2016-09-01 | 2018-03-01 | Alcatel-Lucent Usa Inc. | Estimating bandwidth in a heterogeneous wireless communication system |
US20180092111A1 (en) * | 2016-09-28 | 2018-03-29 | Qualcomm Incorporated | Maximizing a frame's arrangement thereby increasing processing time available to processors |
US10164867B2 (en) * | 2017-02-03 | 2018-12-25 | Cisco Technology, Inc. | Generating non-congruent paths having minimal latency difference in a loop-free routing topology having routing arcs |
Non-Patent Citations (10)
Title |
---|
A. Botta et al., "A Tool for the Generation of Realistic Network Workload for Emerging Networking Scenarios," Computer Networks, 56, pp. 3531-3547 (2012). |
A. Fashad et al., "On the Impact of 802.11n Frame Aggregation on End-To-End Available Bandwidth Estimation," SECON, IEEE, pp. 108-116 (Jun. 2014). |
B.D. Choi., "Priority Queueing System with Fixed-Length Packet-Train Arrivals," IEEE Proceedings Communications, vol. 145, No. 5, pp. 331-336 (Oct. 1998). |
C.D. Mano et al., "RIPPS," ACM Transactions on Information and Systems Security, vol. 11, No. 2, Article 6, pp. 1-23 (May 2008). |
J. Kolap et al., "Frame Aggregation Mechanism for High-Throughput 802.11n Wlans," International Journal of Wireless & Mobile Networks (IJWMN), vol. 4, No. 3, pp. 141-153 (Jun. 2012). |
J. Strauss et al., "A Measurement Study of Available Bandwidth Estimation Tools," IMC '03 ACM Press. |
J. Walraevens et al., "A Discrete-time Priority Queue with Train Arrivals," Stochastic Model, 23, pp. 489-512 (2007). |
S. Hemminger, "Network Emulation with NetEm," Linux Conf Au (Apr. 2005). |
S. Savage, "Sting: A TCP-based Network Measurement Tool," USENIX Symposium on Internet Technologies and Systems, vol. 2, pp. 1-9 (1999). |
V.J. Ribeiro et al., "PathChirp: Efficient Available Bandwidth Estimation for Network Plans," No. SLAC-PUB-9732 (Apr. 2003). |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11025524B2 (en) * | 2018-08-20 | 2021-06-01 | Exfo Inc. | Telecommunications network and services QOE assessment |
Also Published As
Publication number | Publication date |
---|---|
US20180317131A1 (en) | 2018-11-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10390252B2 (en) | User terminal, communication method and apparatus | |
US8472379B2 (en) | Mobile station, radio base station, communication control method, and mobile communication system | |
JP6635044B2 (en) | Radio resource control system, radio base station, relay device, radio resource control method and program | |
JP4799213B2 (en) | Wireless communication terminal and wireless communication method | |
US9736859B2 (en) | Method and apparatus for providing voice service in wireless local area network | |
US20150207700A1 (en) | Evaluation of Radio Network Performance | |
EP2425590B1 (en) | Device and method for computation of channel loss rate and collision loss rate of communication link(s) in a random access network | |
Song et al. | Leveraging frame aggregation for estimating WiFi available bandwidth | |
US9736702B2 (en) | System and method for quality of service control | |
EP3550774A1 (en) | Packet transmission method and hybrid access gateway | |
Slingerland et al. | Performance of transport control protocol over dynamic spectrum access links | |
Dely et al. | BEST-AP: Non-intrusive estimation of available bandwidth and its application for dynamic access point selection | |
US20170041817A1 (en) | Communication management apparatus, wireless terminal, and non-transitory machine-readable storage medium | |
KR102318284B1 (en) | Congestion detection method for data transmission and device therefor | |
US10383002B2 (en) | Systems and methods for rapidly estimating available bandwidth in a WiFi link | |
EP2999268B1 (en) | Wireless fidelity (wi-fi) monitoring method and device | |
CN107371179B (en) | Measurement result reporting method, measurement result receiving method, related equipment and system | |
CN105451337A (en) | Clear channel assessment threshold control method and device | |
EP3354059A1 (en) | REPORTING OF UPLINK (UL) QUALITY OF SERVICE (QoS) METRICS | |
JP6263102B2 (en) | WIRELESS TERMINAL, COMMUNICATION SYSTEM, PROGRAM, AND METHOD WHICH CAN SET WINDOW SIZE ACCORDING TO CONDUCTION LEVEL ON BACKHAUL | |
US8542608B2 (en) | Method for achieving an optimal shaping rate for a new packet flow | |
US9572066B2 (en) | Apparatus and method for access point selection | |
US11303552B2 (en) | Receiver, communication system, available bandwidth estimation method, and storage medium in which program is stored | |
JP4974366B2 (en) | Wireless environment monitoring apparatus and program for monitoring time occupation ratio of wireless environment | |
US20200008197A1 (en) | Communication apparatus, communication method, wireless communication system and program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO SMALL (ORIGINAL EVENT CODE: SMAL); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
AS | Assignment |
Owner name: UNIVERSITY OF NOTRE DAME DU LAC, INDIANA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:STRIEGEL, AARON;SONG, LIXING;REEL/FRAME:045905/0838 Effective date: 20180520 |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
AS | Assignment |
Owner name: NATIONAL SCIENCE FOUNDATION, VIRGINIA Free format text: CONFIRMATORY LICENSE;ASSIGNOR:UNIVERSITY OF NOTRE DAME;REEL/FRAME:054488/0298 Effective date: 20200228 |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2551); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 4 |
|
AS | Assignment |
Owner name: NATIONAL SCIENCE FOUNDATION, VIRGINIA Free format text: CONFIRMATORY LICENSE;ASSIGNOR:UNIVERSITY OF NOTRE DAME;REEL/FRAME:063601/0040 Effective date: 20200228 |